|
As
reported in International Evidence
on Food Consumption Patterns, the budget shares and elasticities
are derived as follows:
International Comparison Project Data
The International Comparison Project (ICP) was originally initiated
by researchers at the University of Pennsylvania and is currently
coordinated by the International Comparison Program, Development
Data Group of the World Bank. The number of countries included in
the ICP data has increased from 10 in 1970 to 115 countries in 1996.
This study uses the 1996 ICP data, which cover expenditure and price
data for 115 countries, 10 broad consumption categories, and 22
subcategories. However, the analysis in the technical bulletin uses
data for only 114 countries, since appropriate population data for
Herzegovina were unavailable.
To conduct cross-country analysis, real consumption expenditures
in different currencies must be expressed in terms of a base-country
currency comparable across countries. One solution is to convert
expenditures into a single currency by using exchange rates. However,
exchange rates do not account for the fact that services are cheaper
in low-income countries. To obtain more accurate estimates for individual
countries, the ICP uses the Geary-Khamis method of aggregation to
obtain prices and volumes in terms of purchasing power parities
relative to a base country. These values allow comparisons at various
levels of aggregation for all countries included in the analysis.
Two-Stage Demand Model
This analysis uses a two-stage demand system. In the first stage,
consumers are assumed to make budget choices over broad consumption
groups. Given the budget endowment for each broad consumption group,
in the second stage consumers make budget choices for consumption
items within each group. In using this approach, we maintain preference
independence within the broad groups. In the second stage, weak
separability among items within a broad group is assumed.
The Florida model, a modified Working's model that incorporates
price terms, is fit to the first-stage model for nine broad groups
of goods across 114 countries. This model assumes preference independence.
A modified version of the Florida model, the Florida Slutsky model,
which assumes weak separability, is fit to the eight food subgroups.
The country data exhibit group heteroskedasticity. A maximum likelihood
procedure that corrects for group heteroskedasticity is developed
and used to estimate the model. Outliers are identified with information
inaccuracy measures, and Strobel measures of goodness-of-fit are
calculated.
The parameters estimated in the first stage of the analysis are
used to calculate elasticities for the broad consumption groups,
and the parameters estimated in the second stage are used to calculate
the conditional elasticities for the food subgroups. The unconditional
elasticities for the food subgroups are then calculated by multiplying
the conditional elasticities by the elasticity for food estimated
in the first stage of the estimation process.
For additional information and references, see International
Evidence on Food Consumption Patterns.
|
